🔥The 9th Round of Easy Loan, Earn $40 Reward is in progress❗️ ⏰ Promotion Period: January 15th - Feburary 15th, 2025 👉 Register now and check more details at gate.io/campaigns/358

At #CVPR2025 today (6/12) and looking forward to give two workshop talks on Mobile Manipulation! 9:30-10am at the MEIS Workshop: Towards Collaborative Mobile Manipulation 2-2:30pm at the OpenSUN3D Workshop: The Challenges and Opportunities of Mobile Manipulation

🌈🤖

Join us at OpenSUN3D☀️ workshop this afternoon #CVPR2025 🚀 📍: Room 105 A 🕰️: 2:00-6:00 pm 🌍: opensun3d.github.io Afshin Dehghan Jeannette Bohg Laura Leal-Taixe

Manifold is now open Exploratorium ! Come visit our giant robot hand that uses AI to learn how to be endearing. Engineering team includes Michelle Pan, Ethan Qiu, Parker Robinson, and Edwin Lai

When you manipulate something, sometimes you just need to get a better view. This robot learns to look. The most interesting thing to me is that it learned to search for objects that are not immediately in view.

Haoyu built an awesome bimanual + neck robot which can be easily mounted on the TidyBot++ mobile base. Hardware design is fully open source! Check out his thread below to learn more 👇

The recordings from our workshop on Open-World 3D Scene Understanding #CVPR2025 are now available! See you #ICCV2025 in Honolulu🏄‍♂️ for the next edition! ➡️ youtube.com/playlist?list=… 🌍 opensun3d.github.io

Good morning LA 🌞#RSS2025 Looking forward to give a workshop talk today on all the recent progress we have made in Mobile Manipulation!

I want to highlight one important result: Data curated with CUPID 💘 for single-task diffusion policies improves post-training performance of VLAs like π0. We achieve this with influence functions that let us trace back from policy performance to training demonstrations.

We use a VLM as a high-level planner for dexterous manipulation and then achieve fine control via residual RL. Check out Vincent de Bakker thread for more details.

Excited to share CUPID 💘 won the Best Paper Award at the #RSS2025 RoboEval workshop! TL;DR: In 🤖 imitation learning, data quality is crucial. CUPID 💘 directly identifies “good” robot data, i.e., whether a demo will improve success rates. ➡️ big gains via data curation! 🧵👇